Geomagnetic navigation algorithm: A novel meta-heuristic algorithm for solving engineering applications

This paper proposes a novel metaheuristic optimization algorithm-Geomagnetic Navigation Algorithm (GNA).inspired by the migratory behavior of birds, which navigate using geomagnetic inclination, calibrate with visual landmarks, and adapt to turbulent disturbances. The algorithm introduces an innovative threefold mathematical model: (1) a navigation term constructed by mapping the solution space gradient through geomagnetic inclination; (2) a landmark-driven cooperative convergence strategy based on dynamically selecting the top three individuals in the population; and (3) a hybrid perturbation mechanism combining Cauchy, Gaussian, and Levy flights in a 1:1:1 ratio to enhance exploration capabilities.Experimental evaluations on the CEC-2017 benchmark functions in 10-, 30-, and 50-dimensional settings, as well as real-world engineering problems such as pressure vessel design, demonstrate that GNA outperforms six state-of-the-art algorithms including WOA, TOC, and PO. The results confirm that GNA achieves superior performance in exploration.exploitation balance, resistance to local optima, scalability with dimensionality, and constraint-handling efficiency, offering an effective solution for complex optimization problems.